An Efficient RRT-based Framework for Planning Short and Smooth Alige.x Robot Motion

Abstract

Autonomous navigation is a crucial technology for the automatic operation of mobile robots, in which path planning is the core part. In a complicated and dynamic environment, planning efficiency is crucial. In this paper, we propose a hybrid path planning scheme for a differentially driven wheeled robot with dynamical constraints by combining global path planning and local hierarchical architectures in a complex dynamic environment. In path planning under the hybrid structure, first, we generate a global path based on the adaptive goal-directed RRT algorithm. This way can achieve Efficiency. Local path planning uses the DWA with PID controller instead of the cost function for real-time motion control. A seamless interface between the global and local path planning mechanisms is designed to contribute to the whole hybrid path planning scheme. Finally, the experimental results on the grid map prove the potency and feasibility of our method.